Dental Sonorants or Approximants (Re: Terms for "white man")

[Koontz John E]

On Wed, 17 Mar 2004, Alan Hartley wrote:
> There were Ojibway dialects at various times and places with reflexes r,
> l, and n for Proto-Algonquian l; it's n in modern O.)
>
Rankin:
> > Might have something to do with the fact that Cree, like Siouan, has the
> > [dh, n, r, l, y] variation for its dental approximant phoneme.

It seems like this is characteristic of all of the large Northeastern
dialect chains.  In Siouan we might say that we have two or three of these
chains, e.g., Dakotan (Te l ~ n, Sa d ~ n, Ya d ~ n, As n, St n - and
really these all alternate with y in historically unclustered contexts,
too), and Dhegiha (OP dh ~ n, Ks y ~ l ~ n, Os dh ~ r ~ l ~ n, Qu d ~ n,
with the first alternant in Ks and OS being the unclustered alternant).
Winnebago and Ioway-Otoe are a bit too far apart to be called dialects,
but Winnebago has r ~ n, and IO has something that is usually written r ~
n at the moment, but has been written l ~ n in the past.  In all cases
where there is an n alternant, this is what occurs before nasal vowels,
albeit not all nasal vowels, and some n's seem to have a different source.
If there's only an n variant, then n occurs everywhere.  (However, I'm not
completely positive this is a fair characterization of Assiniboine or
Stoney.)

Outside of Mississippi Valley Siouan, in Missouri River Siouan or
Crow-Hidatsa, Kaschube writes r for Crow (perhaps an arbitrary choice of a
neutral symbol for the underlying entity), while the current popular
orthography writes d ~ l ~ n, depending on context.  Hidatsa specialists
have somewhat similar ranges of practices, but I think at present they are
writing n ~ r, depending on context.  These languages don't have nasal
vowels, and Crow n occurs in rr clusters (nn) and, I think finally, while
in Hidatsa it occurs at least initially and probably also in clusters,
though I don't recall the details.